Share Email Print

Proceedings Paper

Unsupervised segmentation of 3D medical images based on clustering and deep representation learning
Author(s): Takayasu Moriya; Holger R. Roth; Shota Nakamura; Hirohisa Oda; Kai Nagara; Masahiro Oda; Kensaku Mori
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper presents a novel unsupervised segmentation method for 3D medical images. Convolutional neural networks (CNNs) have brought significant advances in image segmentation. However, most of the recent methods rely on supervised learning, which requires large amounts of manually annotated data. Thus, it is challenging for these methods to cope with the growing amount of medical images. This paper proposes a unified approach to unsupervised deep representation learning and clustering for segmentation. Our proposed method consists of two phases. In the first phase, we learn deep feature representations of training patches from a target image using joint unsupervised learning (JULE) that alternately clusters representations generated by a CNN and updates the CNN parameters using cluster labels as supervisory signals. We extend JULE to 3D medical images by utilizing 3D convolutions throughout the CNN architecture. In the second phase, we apply k-means to the deep representations from the trained CNN and then project cluster labels to the target image in order to obtain the fully segmented image. We evaluated our methods on three images of lung cancer specimens scanned with micro-computed tomography (micro-CT). The automatic segmentation of pathological regions in micro-CT could further contribute to the pathological examination process. Hence, we aim to automatically divide each image into the regions of invasive carcinoma, noninvasive carcinoma, and normal tissue. Our experiments show the potential abilities of unsupervised deep representation learning for medical image segmentation.

Paper Details

Date Published: 12 March 2018
PDF: 7 pages
Proc. SPIE 10578, Medical Imaging 2018: Biomedical Applications in Molecular, Structural, and Functional Imaging, 1057820 (12 March 2018); doi: 10.1117/12.2293414
Show Author Affiliations
Takayasu Moriya, Nagoya Univ. (Japan)
Holger R. Roth, Nagoya Univ. (Japan)
Shota Nakamura, Nagoya Univ. Graduate School of Medicine (Japan)
Hirohisa Oda, Nagoya Univ. (Japan)
Kai Nagara, Nagoya Univ. (Japan)
Masahiro Oda, Nagoya Univ. (Japan)
Kensaku Mori, Nagoya Univ. (Japan)

Published in SPIE Proceedings Vol. 10578:
Medical Imaging 2018: Biomedical Applications in Molecular, Structural, and Functional Imaging
Barjor Gimi; Andrzej Krol, Editor(s)

© SPIE. Terms of Use
Back to Top